The proposed experiments will advance our understanding of how aging affects language production by using dual task procedures to assess the costs of language production to young and older adults. Moment-to-moment accuracy of pursuit-rotor tracking will be studied as an indicator of moment-to-moment processing demands of language production. Experiment 1 will establish baselines for tracking accuracy and compare task priorities of young and older adults engaged in rotor tracking while orally responding to thought-provoking questions. In Experiment 2, young and older adults'tracking accuracy will be monitored as they engage in a variety of language production tasks ranging from rote repetition of the alphabet, to counting backwards, to reading sentences and prose, and spontaneous speech (responding to thought-provoking questions). In order to provide a more fine grained assessment of the moment-to-moment cognitive costs of language production, in Experiments 3 and 4, continuous records of tracking accuracy will be synchronized and time-locked to a continuous record of language production while young and older adults speak or read aloud. The speech signal will be segmented to mark the onset/offset of sentences and clauses. Average tracking accuracy, will then be computed for these segments, yielding time-on-task measurement of tracking accuracy that correspond to critical speech "events" such as different types of embedded clauses. Average tracking accuracy is expected to decrease with sentence and clause complexity. In Experiment 5, tracking accuracy will be monitored while participants engage in a controlled production task that separates sentence planning from plan implementation and sentence production. The central hypothesis is that language production is costly to older adults who must draw upon cognitive reserve capacity in order to respond to dual task demands. In demanding situations, e.g., producing a complex sentence, older adults'cognitive reserve will be insufficient. As a result, their control over tracking will deteriorate, resulting in decreased accuracy. As linguistic task demands increase, older adults'speech will erode, first showing a reduction in speed, then a loss of fluency more generally, and finally a loss of grammatical complexity and propositional content.